Chi-Hua Chien was a 27-year-old associate at Accel when he first encountered a six-person company launched from Harvard called The Facebook. More than two decades later, as co-founder of Goodwater Capital, he is applying that same ability to read human behavior at scale to the AI boom — and his conclusions challenge the prevailing narrative.
In a recent interview, Chien argued that the commoditization of the AI model layer is already underway. He pointed to Google’s decision to drop the price of its subscription AI product from $7.99 to $4.99 per month while doubling storage as evidence of price competition intensifying among infrastructure players.
Also read: Qualcomm unveils Snapdragon Reality Elite and START toolkit to power the next wave of AI wearables
Following the pattern of past tech cycles
Chien draws on historical data to support his thesis. In the web era, he noted, infrastructure new entrants produced $400 billion in new market cap, while application companies created $3.1 trillion — 88% of the new value. During the mobile cycle, the pattern held: infrastructure generated about $700 billion versus $3.7 trillion for applications, including companies like Netflix, Spotify, Meta, Uber, and Airbnb.
“Infrastructure market caps actually peaked in the year 2000,” Chien said. “But you fast forward 25, 26 years later, and in nominal dollar terms, the market cap of those infrastructure companies has not surpassed the 2000 peak.”
Also read: The US government's Anthropic models ban was never about an AI jailbreak
Hyper-personalization as the differentiator
Chien sees personalization as the through line separating the next wave of winners. He pointed to portfolio companies like Triumph, Ritten, and Flow GPT, which are generating hundreds of millions in annual recurring revenue by using AI to make experiences more customizable — without marketing themselves as AI companies.
“The customer is not saying ‘this is an AI application,'” Chien said. “They’re saying it’s an entertainment application.”
In healthcare, he cited Midi Health, a women’s health company that uses AI to expand the supply of care for perimenopausal women, treating hundreds of thousands of patients in a market previously constrained by a shortage of trained providers.
The shrinking gap between local and frontier AI
Chien also highlighted a technical trend with significant implications: the gap between what can run locally on a phone and the most advanced cloud-based models is shrinking rapidly. Two years ago, the lag was 18 to 24 months. It is now six months, and Chien expects it to reach three months within the next year.
“What we don’t yet have is the use cases very well defined,” he said, drawing a parallel to the early iPhone era, when many assumed mobile would simply be a port of web applications.
Why a US super app remains elusive
Chien also addressed why Facebook and other tech giants have failed to build a super app combining social and financial services for American consumers. He cited a fundamental trust gap: financial transactions require high confidence in security and reliability, while social media thrives on triviality.
“Financial services transactions are very high monetization and relatively low time,” Chien said. “You don’t want to hang out in your banking app. You want to transact and be done.”
Betting on in-person connection
Chien is also investing in the countertrend to digital saturation. Goodwater has backed companies like Bump, founded by the original creators of Zenly, and Fever, often described as the Live Nation of Europe. Both focus on using digital tools to catalyze real-world experiences.
“What do people crave in a world where there’s an infinite supply of digital content?” Chien asked. “They crave the thing that is most constrained, which is real human contact, real-world experiences.”
Frequently Asked Questions
What is Goodwater Capital?
Goodwater Capital is a venture capital firm co-founded by Chi-Hua Chien that focuses exclusively on consumer and prosumer technology investments across entertainment, healthcare, fintech, and live experiences.
How did Chien find Facebook?
As a 27-year-old associate at Accel in 2005, Chien identified the six-person company launched from Harvard and helped lead Accel’s early investment in what became one of the most successful venture bets in history.
What does Chien mean by ‘model layer commoditization’?
He argues that the underlying AI models themselves will become interchangeable commodities, with value accruing to companies that build applications using those models to deliver personalized user experiences.
